摘要
由一名新闻记者-机器人与机器学习的工作人员新闻编辑每日新闻-调查人员发布了关于马学习的新报告。根据NewsRx记者从法国圣保罗Les Durance发来的消息，研究称，在C倾斜和退役操作(MICADO)H2020项目的测量和仪器项目中，开发了一个被动和主动中子测量系统，以估计低和中等放射性水平的遗留废桶内的净物质质量。进行了蒙特卡洛模拟，设计了一个可运输的中子系统，允许被动中子符合计数和使用差分消失技术(DDT)的主动交叉。这项研究的财政支持来自地平线2020.我们的新闻记者从CEA的研究中获得了一句话，“然而，在这两种测量模式中代表感兴趣信号(由于核Ar材料)的校准系数(CCs)可能会随核废料桶基质的性质而变化很大。因此，本文基于104个Monte-Carlo计算，用不同的废料桶研究基质效应。”基于田口实验设计，具有一系列密度、材料组成、填充水平、用机器学习算法研究了矩阵校正方法,从位于测量腔内的内中子监测器信号和AmBe中子源透射测量中,推导出矩阵对中子信号的影响,这些量可以通过实验评估,并作为解释变量,用于定义simu预测模型。建立了基于普通最小二乘法(OLS)的多线性回归模型,并与随机森林(RF)机器学习算法和多层感知器(MLP)人工神经网络进行了比较。MLP法和RF法预测的CS值分别优于17%和3%,而OLS法预测的CS值均在95%置信水平范围之外,MLP法和RF法预测的CS值与OLS法预测的CS值的一致性分别优于17%和3%。得出了类似的结论.四个模型滚筒的CC预测优于相应的MLP、RF和OLS方法的12%、35%和72%.
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Investigators publish new report on Ma chine Learning. According to news originating from St. Paul les Durance, France, by NewsRx correspondents, research stated, "A passive and active neutron measur ement system has been developed within the Measurement and Instrumentation for C leaning and Decommissioning Operation (MICADO) H2020 project to estimate the nuc lear material mass inside legacy waste drums of low and intermediate radioactivi ty levels. Monte-Carlo simulations were performed to design a transportable neut ron system allowing both passive neutron coincidence counting and active interro gation with the differential die-away technique (DDT)." Financial support for this research came from Horizon 2020. Our news journalists obtained a quote from the research from CEA, "However, the calibration coefficients (CCs) representing the signal of interest (due to nucle ar material) in these two measurement modes may vary by a large amount depending on the properties of the matrix of the nuclear waste drum. Therefore, this arti cle investigates matrix effects based on 104 Monte-Carlo calculations with diffe rent waste drums, based on Taguchi experimental design with a range of densities , material compositions, filling levels, and nuclear material masses. A matrix c orrection method is studied using machine learning algorithms. The matrix effect on the neutron signal is deduced from the signal of internal neutron monitors l ocated inside the measurement cavity and from a transmission measurement with an AmBe neutron source. Those quantities can be assessed experimentally and are us ed as explanatory variables for the definition of a predictive model of the simu lated CC, either in passive or in active mode. A multilinear regression model of the CC based on ordinary least square (OLS) is built and compared to the random forest (RF) machine-learning algorithm and to the multilayer perceptron (MLP) a rtificial neural network. In passive neutron coincidence counting, the residual error of the regression is lower for the MLP and RF than for OLS. The agreement between the predicted CCs of four mockup drums used as test is better than 17% and 3%, respectively, with the MLP and RF methods, while three pred ictions are out of the 95 % confidence level range with OLS. In act ive neutron interrogation, similar conclusions are drawn. The prediction of the CC for the four mockup drums is better than 12%, 35%, and 72% for the respective MLP, RF, and OLS methods."
NETL
NSTL